Trump administration demands OpenAI restrict GPT-5 release

The Trump administration has taken an unprecedented step by ordering OpenAI to restrict the rollout of its next major AI model, reportedly GPT-5. Under the directive, the company must allow the government to vet the first 5.6 users before a wider launch, with each initial partner requiring government approval. This marks the first time a US AI firm has been compelled to limit a model's release, highlighting growing federal oversight of frontier AI development.

Meanwhile, Anthropic remains locked in its own dispute with Washington over model safety and transparency. The geopolitical implications are stark: as AI investor Nathan Benaich noted, 'what the rest of us are permitted to do with it can change on a Friday afternoon.' This move signals a shift toward tighter controls on AI deployment, potentially reshaping global access to cutting-edge models.